NAVER AI LAB (@NAVER_AI_Lab) profile picture

NAVER AI LAB

@NAVER_AI_Lab
Joined November 2021
  • Tweets34
  • Following1
  • Followers638
2Media
Photo by NAVER AI LAB (@NAVER_AI_Lab)
Please check out the presentation schedule of #NAVER #CLOVA and #AI_Lab (@ClovaAiLab) at #ICLR2022 (@iclr_conf): 12 papers in the main session. My sessions are posters-3, 4, and 7. Also, please visit ICLR 2022 Social: ML in Korea (Apr 28, 12-3PM, KST) 👉naver-career.gitbook.io/en/t…
Image from 국가AI전략위 부위원장 하정우 (@JungWooHa2)
Image from 국가AI전략위 부위원장 하정우 (@JungWooHa2)
6
30
Happy to share #CRAFT proposal (#HyperscaleFAccT) of our #CLOVA was accepted to @FAccTConference 2022! HyperscaleFAccT addresses fairness, accountability, and transparency in hyperscale LM (#HyperCLOVA). Thanks to all contributors; Hwaran, @aliceoh @mgalle @nekozzang @kchonyc
Image from 국가AI전략위 부위원장 하정우 (@JungWooHa2)
2
25
So excited to share that #NAVER #CLOVA (@ClovaAiLab ) has 14 papers accepted at #cvpr2022 (@CVPR ). Congrats to all the authors including @JihwanBang, @SanghyukChun , @coallaoh , @ppolon , and @yunjey_choi ! I wish we meet all in June in New Orleans !
Image from 국가AI전략위 부위원장 하정우 (@JungWooHa2)
15
4
83
[1/12] Learning Features with Parameter-Free Layers Dongyoon Han (NAVER AI Lab), YoungJoon Yoo (NAVER Clova), Beomyoung Kim (NAVER Clova), Byeongho Heo (NAVER AI Lab) openreview.net/forum?id=bCrd…
3
4
[2/12] SUMNAS: Supernet with Unbiased Meta-Features for Neural Architecture Search. Hyeonmin Ha (SNU), Ji-Hoon Kim(NAVER Clova), Semin Park (Google), Byung-Gon Chun (SNU) openreview.net/forum?id=Z8Fz…
2
2
[3/12] ViDT: An Efficient and Effective Fully Transformer-based Object Detector. Hwanjun Song (NAVER AI Lab), Sanghyuk Chun (NAVER AI Lab), Dongyoon Han (NAVER AI Lab), Byeongho Heo (NAVER AI Lab), Wonjae Kim (NAVER AI Lab) et al. openreview.net/forum?id=w4cX…
3
4
[4/12] Coherence-based Label Propagation over Time Series for Accelerated Active Learning. Yooju Shin (KAIST), Susik Yoon (UIUC), Sundong Kim (IBS), Hwanjun Song (NAVER AI Lab), Jae-Gil Lee (KAIST), Byung Suk Lee (Univ. of Vermont) openreview.net/forum?id=gjNc…
2
2
[5/12] Which Shortcut Cues Will DNNs Choose? A Study from the Parameter-Space Perspective. Luca Scimeca (NAVER AI Lab), Seong Joon Oh (NAVER AI Lab), Sanghyuk Chun (NAVER AI Lab), Michael Poli (NAVER AI Lab), Sangdoo Yun (NAVER AI Lab) openreview.net/forum?id=qRDQ…
3
3
[6/12] Know Your Action Set: Learning Action Relations for Reinforcement Learning. Ayush Jain*(USC), Norio Kosaka*(NAVER CLOVA), Kyung-Min Kim(NAVER CLOVA), Joseph J Lim(KAIST, USC, & NAVER AI Lab) openreview.net/forum?id=MljX…
2
5
[7/12] Skill-based Meta-Reinforcement Learning Taewook Nam (KAIST), Shao-Hua Sun (USC), Karl Pertsch (USC), Sung Ju Hwang (KAIST), Joseph J Lim(KAIST, USC, & NAVER AI Lab) openreview.net/forum?id=jeLW…
2
2
[8/12] Task-Induced Representation Learning. Jun Yamada (Oxford), Karl Pertsch (USC), Anisha Gunjal (USC), Joseph J Lim(KAIST, USC & NAVER AI Lab) openreview.net/forum?id=OzyX…
5
5
[9/12] Encoding Weights of Irregular Sparsity for Fixed-to-Fixed Model Compression Baesung Park*(NAVER CLOVA), Sejung Kwon*(NAVER CLOVA), Daehwan Oh (Samgsung Research), Byeongwook Kim (NAVER CLOVA), Dongsoo Lee (NAVER CLOVA) openreview.net/forum?id=Vs5N…
2
4
[10/12] Online Continual Learning on Class Incremental Blurry Task Configuration with Anytime Inference. Hyunseo Koh (GIST, NAVER AI Lab), Dahyun Kim (GIST, NAVER AI Lab), Jung-Woo Ha (NAVER AI Lab), Jonghyun Choi (GIST, NAVER AI Lab) openreview.net/forum?id=nrGG…
2
2
[11/12] Generating Videos with Dynamics-aware Implicit Generative Adversarial Networks Sihyun Yu (KAIST), Jihoon Tack (KAIST), Sangwoo Mo (KAIST), Hyunsu Kim (NAVER AI Lab), Junho Kim (NAVER AI Lab), Jung-Woo Ha (NAVER AI Lab), Jinwoo Shin (KAIST) openreview.net/forum?id=Czsd…
2
3
[12/12] Contrastive Fine-grained Class Clustering via Generative Adversarial Networks. Yunji Kim (NAVER AI Lab), Jung-Woo Ha (NAVER AI Lab) openreview.net/forum?id=XWOD…
2
10
[1/7] MULTI-SCALE SPEAKER EMBEDDING-BASED GRAPH ATTENTION NETWORKS FOR SPEAKER DIARISATION Youngki Kwon (NAVER CLOVA), Hee-Soo Heo (NAVER CLOVA), Jee-weon Jung (NAVER CLOVA), You Jin Kim (NAVER CLOVA), Bong-Jin Lee(NAVER CLOVA), Joon Son Chung (KAIST) arxiv.org/abs/2110.03361
Multi-scale speaker embedding-based graph attention networks for...

Multi-scale speaker embedding-based graph attention networks for...

The objective of this work is effective speaker diarisation using multi-scale speaker embeddings. Typically, there is a trade-off between the ability to recognise short speaker segments and the...

arxiv.org
1
3
[2/7] SPELL MY NAME: KEYWORD BOOSTED SPEECH RECOGNITION Namkyu Jung (NAVER CLOVA), Geonmin Kim (NAVER CLOVA), Joon Son Chung (KAIST) arxiv.org/abs/2110.02791
Spell my name: keyword boosted speech recognition

Spell my name: keyword boosted speech recognition

Recognition of uncommon words such as names and technical terminology is important to understanding conversations in context. However, the ability to recognise such words remains a challenge in...

arxiv.org
1
2
[3/7] INTEGRATION OF PRE-TRAINED NETWORKS WITH CONTINUOUS TOKEN INTERFACE FOR END-TO-END SPOKEN LANGUAGE UNDERSTANDING Seunghyun Seo (NAVER CLOVA), Donghyun Kwak (NAVER CLOVA), Bowon Lee (Inha U.) arxiv.org/abs/2104.07253
Integration of Pre-trained Networks with Continuous Token...

Integration of Pre-trained Networks with Continuous Token...

Most End-to-End (E2E) SLU networks leverage the pre-trained ASR networks but still lack the capability to understand the semantics of utterances, crucial for the SLU task. To solve this, recently...

arxiv.org
1
2
[4/7] AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks J.-W Jung (CLOVA), H.-S. Heo (CLOVA), H. Tak (EURECOM), H.-J. Shim (U. of Seoul), J. S. Chung (KAIST), B.-J. Lee (CLOVA), H.-J. Yu(U. of Seoul), N. Evans(EURECOM) arxiv.org/abs/2110.01200
AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal...

AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal...

Artefacts that differentiate spoofed from bona-fide utterances can reside in spectral or temporal domains. Their reliable detection usually depends upon computationally demanding ensemble systems...

arxiv.org
1
3
[5/7] ATTENTIVE MAX FEATURE MAP AND JOINT TRAINING FOR ACOUSTIC SCENE CLASSIFICATION Hye-jin Shim (U. of Seoul), Jee-weon Jung (NAVER CLOVA), Ju-ho Kim (U. of Seoul), Ha-Jin Yu (U. of Seoul) arxiv.org/abs/2104.07213
Attentive max feature map and joint training for acoustic scene...

Attentive max feature map and joint training for acoustic scene...

Various attention mechanisms are being widely applied to acoustic scene classification. However, we empirically found that the attention mechanism can excessively discard potentially valuable...

arxiv.org
1
1
Load more